This proposal addresses a key barrier to research in nutritional epidemiology: the small number of objective biomarkers available for assessing relationships between diet and health. Because diet-disease associations may be subtle and are difficult to establish using self-reported methods, there is a pressing need to develop and validate more objective dietary biomarkers for use in research. The proposed study will evaluate a set of dietary biomarkers for sweetener, meat and fish intake based on the stable isotope ratios of carbon, nitrogen and sulfur. The carbon isotope ratio is a candidate biomarker of sweetener intake because it is high in corn and sugar cane, the major sources of dietary sweeteners in the US. The nitrogen isotope ratio is a candidate biomarker of meat intake, because it is elevated in animals relative to plants. The sulfur isotope ratio is a candidate biomarker of fish intake, because marine foods are elevated relative to terrestrial foods. However, these measures have not yet been validated in an appropriate framework, which has prevented their adoption as tools for nutritional epidemiology. Here we test two approaches to validating isotopic measures for foods of interest: first, by using multi- isotopic models of diet o increase specificity, and second, by measuring the stable isotope ratios of specific amino acids that are closely linked to those foods. We test these approaches using serum specimens from the NPAAS Feeding Study, a 2-week controlled feeding study of 150 women ages 60-80 in Seattle, WA. The NPAAS Feeding Study individually matches each participant's controlled diet to her usual intake, in order to maintain intake distributions that are similar to those of the stuy population and achieve the level of control needed for biomarker validation. Specific Aim 1 will test whether single or multi-isotopic models based on serum carbon, nitrogen, and sulfur stable isotope ratios can provide valid and specific measures of sweetener, fish and meat intake in the NPAAS Feeding Study. Specific Aim 2 will test whether the carbon isotope ratio of specific serum amino acids can provide alternatives to multi-isotopic models of similar or improved validity and specificity. All stable isotope ratios will be measured using continuous-flow isotope ratio mass spectrometry. This proposal is a first step toward validating multiple stable isotope biomarkers of diet as new tools for objective dietary assessment. Using samples from the NPAAS Feeding Study provides an efficient and cost-effective opportunity to critically assess and refine these markers for more general use in epidemiologic research.